Skip to content

Instantly share code, notes, and snippets.

View nikkipin's full-sized avatar

Nikki Pinski nikkipin

  • Safety Cyber Security
  • Vancouver
View GitHub Profile
@nikkipin
nikkipin / SKILL.md
Created September 9, 2026 16:23
/eli5 skill for claude
name eli5
description Use when explaining a complex technical concept, system, decision, or piece of code in plain English to someone smart but new to the domain — in conversation or in writing. Triggers include "eli5 X", "explain like I'm five", "I don't understand X", "walk me through this", "what does this actually do", "you've lost me", "say it straight", "in plain English", "drop the jargon", or any explanation where the audience shouldn't be assumed to know the jargon.

/eli5

Never open with a content-free lead-in. No "The one idea…", "The big idea…", "The one thing…", "Here's the gist…", or any paraphrase that announces an insight is coming instead of stating it. This ban covers the first words out — conversational reply or heading alike — not only section titles. If the explanation has a load-bearing insight, open with the insight itself, named for its content ("Fix commits are our entire universe"); if it doesn't, open with the problem. See "Lead with the load-bearing insight" belo

@nikkipin
nikkipin / CLAUDE.md
Created May 11, 2026 17:34
karpathy+mnimy

CLAUDE.md — N-rule template

These rules apply to every task in this project unless explicitly overridden. Bias: caution over speed on non-trivial work. Use judgment on trivial tasks.

Rule 1 — Think Before Coding

State assumptions explicitly. If uncertain, ask rather than guess. Present multiple interpretations when ambiguity exists. Push back when a simpler approach exists. Stop when confused. Name what's unclear.

@nikkipin
nikkipin / SKILL.md
Last active August 10, 2026 21:11
/readable-python (v0.3)
name readable-python
description Use when writing or modifying any Python code, including small local edits, single-line bug fixes, refactors, and new features. Applies on every Python change, not only large restructures.

Readable Python

Readability is attention allocation. Code has foreground (domain logic) and background (plumbing). Make foreground prominent and background invisible, so the reader spends cognitive load only on what's load-bearing.

Rule scope